For nearly a half century, epidemiologists have been intrigued by international and racial-ethnic variations in cancer risk, including cancers of the breast, prostate, and colorectum. The etiology of these cancers is most likely due to a complex interplay between genetic and environmental risk factors. To explore these factors, we established a multi-ethnic cohort of 215,251 men and women in Los Angeles and Hawaii (MEC) in 1993, to study environmental, especially dietary, determinants of risk. The current grant began in 1996 to evaluate the genetic component of racial/ethnic variability in risk to these cancers and the interplay between genes and environment. Using a candidate gene approach we have made significant progress in terms of understanding the genetic basis of these diseases in the past four years, even though we are still in the early stages of understanding the molecular genetic pathways which cause these diseases (see Progress Report). To expand these gene environment studies to include more rapidly fatal cancers of these and other sites, we propose in this renewal application to establish a biorepository of prospectively collected blood specimens from 46,000 African American, Latino, and Japanese participants in the Los Angeles portion of the MEC, ultimately resulting in 100,000 samples in the combined Los Angeles and Hawaii biorepository. We believe that beyond our initial studies, the MEC has tremendous potential to serve as a national resource for studying genetic and environmental causes of cancer and other chronic disease endpoints that are important causes of morbidity and mortality in the minority populations of the MEC and nationally. Their special lifestyle and cultural characteristics and their unusual distribution of cancer and mortality outcomes make these populations particularly interesting for gene by environment studies. We have developed a highly efficient system of sample collection to minimize the substantial cost involved in this undertaking. In this renewal application we also propose to continue our exploration of molecular genetic pathways using both the candidate gene hypothesis driven approach as we have in the past four years, but also using a high-throughput genome analysis to identify and evaluate novel candidate gene polymorphisms through an ongoing collaboration with the Whitehead/MIT Cancer for Genome Research, a leading center for genomic and human genetics.